Inferring multimodal itineraries from rich smartphone data

David Montoya 1, 2 Serge Abiteboul 1, 3
1 DAHU - Verification in databases
LSV - Laboratoire Spécification et Vérification [Cachan], ENS Cachan - École normale supérieure - Cachan, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8643
Abstract : We designed a system to infer the multimodal itineraries traveled by a user from a combination of smartphone sensor data (e.g., GPS, Wi-Fi, inertial sensors), personal information , and knowledge of the transport network topology (e.g., maps, transportation timetables). The system operates with a Multimodal Transport Network that captures the set of admissible multimodal itineraries, i.e., paths of this network with weights providing the statistics (expected time and variance) of the paths. The network takes into account public transportation schedules. Our Multimodal Transport Network is constructed from publicly available transport data of Paris and its neighbourhoods published by different transport agencies and map organizations. The system models sensor uncertainty with probabilities, and the likelihood that a multimodal itinerary was taken by the user is captured in a Dynamic Bayesian Network. For this demonstration, we captured data from users travelling over the Paris region who were asked to record data for different trips via an Android application. After uploading their data into our system, a set of most likely itineraries is computed for each trip. For each trip, the system displays recognized multimodal itineraries and their estimated likelihood over an interactive map.
Type de document :
Communication dans un congrès
BDA 2014 - Gestion de données - principes, technologies et applications, Oct 2014, Autrans, France. pp.38--42, 2014
Liste complète des métadonnées

Littérature citée [28 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01163802
Contributeur : David Gross-Amblard <>
Soumis le : lundi 15 juin 2015 - 15:30:47
Dernière modification le : jeudi 7 février 2019 - 17:29:31
Document(s) archivé(s) le : mardi 25 avril 2017 - 08:14:43

Fichier

bda2014-actes-demo-2-pp38-42.p...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité - Pas d'utilisation commerciale - Pas de modification 4.0 International License

Identifiants

  • HAL Id : hal-01163802, version 1

Collections

Citation

David Montoya, Serge Abiteboul. Inferring multimodal itineraries from rich smartphone data. BDA 2014 - Gestion de données - principes, technologies et applications, Oct 2014, Autrans, France. pp.38--42, 2014. 〈hal-01163802〉

Partager

Métriques

Consultations de la notice

317

Téléchargements de fichiers

210